Skip to content

Waveform library

The library connects a model's usable waveform, its published definition, and the checks actually performed. Start here when you want to know which equations a DiffGW model uses or what a mathematical-reference result covers. For every implemented model, including those without a versioned mathematical reference, use the full dispatch and specification coverage and the broader catalog and support guide. Provisional entries in the coverage table link to source-backed research ledgers. They identify the equations, native components, and missing evidence needed before an executable reference can be trusted. Former complete-waveform prototypes have a separate historical record; they do not appear in the current public dispatch inventory. The qualification status review separates supported implementation scopes from concrete remaining blockers and physical failures.

Model and specification Definition version Scope Backends
TaylorF2 taylorf2-nonspinning-3.5pn/v1 Nonspinning 3.5PN inspiral with restricted quadrupole amplitude and positive frequencies. torch, jax
TaylorF2_Eccentric taylorf2-eccentric-nonspinning-e2-3pn/v1 Nonspinning comparable-mass (q <= 2), 0 < e0 <= 0.1, positive-frequency inspiral: restricted quadrupole amplitude and 3PN order-e0² eccentric phase at explicit f0_ecc and f_ref, with amplitude correction off. torch, jax
TaylorF2_HM taylorf2-hm-leading-carrier/v1 Eight leading nonspinning SPA mode amplitudes with inherited aligned-spin TaylorF2 phase and negative-m polarization projection. torch, jax
IMRPhenomD imrphenomd-intrinsic-composition/v1 Aligned-spin (2,2) IMR composition with declared fit rows and QNM data. torch, jax
IMRPhenomHM imrphenomhm-independent-waveform/v3 Bounded independent mathematical composition for published D calibration and named source PN variants; inherited defaults remain separate. torch, jax
IMRPhenomXAS imrphenomxas-composed-polarizations-carrier/v1 Aligned-spin (2,2) carrier composition with supplied fit and QNM outputs. torch, jax
KerrRingdown220Direct kerr-ringdown-220-energy-direct-class/v1 Direct Torch/JAX KerrRingdown class only: one physical-energy spheroidal (2,2,0) mode with independent QNM; selected FP64 inputs. torch, jax
ContinuousWaveH1Direct cw-monochromatic-h1-direct-class/v2 Direct Torch/JAX ContinuousWave class: selected float64 CPU monochromatic principal-axis 2f strain at H1, GPS epoch in 2015–2016. torch, jax
MemoryWaveform22Direct memory-dominant-22-direct-class/v1 Only Torch/JAX MemoryWaveform.evaluate_22 with supplied physical-strain h22, symmetric mirror, and relative memory offset on selected CPU float64 grids. torch, jax
taylorf2-aligned-bh-spin-3.5pn/v1 taylorf2-aligned-bh-spin-3.5pn/v1 Aligned black-hole spin phase corrections through 3.5PN conservative dynamics and flux to infinity with restricted quadrupole amplitude. torch, jax
taylorf2-adiabatic-quadrupolar-tides-6.5pn/v1 taylorf2-adiabatic-quadrupolar-tides-6.5pn/v1 Adiabatic quadrupolar gravitoelectric TaylorF2 tidal phase corrections through 6.5PN added to aligned-spin TaylorF2. torch, jax
taylorf2-spin-induced-quadrupole-3.5pn/v1 taylorf2-spin-induced-quadrupole-3.5pn/v1 Spin-induced quadrupole moment phase corrections through 3.5PN added to aligned-spin and tidal TaylorF2. torch, jax
ppe-supported-equations/v2 ppe-supported-equations/v2 Parameterized post-Einstein inspiral amplitude modification, phase increment, and massive-graviton dispersion applied to supported GR carriers. torch, jax
emri.newtonian_eccentric_quadrupole.v1 emri-newtonian-eccentric-quadrupole/v1 Weak-field, leading-quadrupole Newtonian eccentric EMRI polarizations at a fixed observer with adiabatic Peters p,e evolution. torch, jax

Each model link opens its canonical specification with equations, component IDs, source versions, variants, and evidence. The TaylorF2 entry covers the nonspinning base; its separately versioned spin, tidal, and quadrupole extensions are linked from that specification. The component specification index lists individual components and extensions. The generation guide explains how to evaluate a waveform; the validation overview records numerical comparisons and their limits.

These are different claims: a cited equation needs source review; agreement with an independent mathematical reference checks an implementation on declared cases; agreement with LALSimulation and accuracy against physical data require their own evidence. An entry in this library is not a blanket accuracy rating.

Python users can inspect the same model and component relationships through diffgw.waveforms, without installing the optional SymPy reference dependencies.